added progress instead of spinner

app.py

@@ -53,8 +53,12 @@ def convert_speech(filename, progress=gr.Progress()):
     with torch.inference_mode():
         # Extract speech units
         units = hubert_loaded.units(source)
+        progress(0.5, desc="Generating target spectrogram")
+
         # Generate target spectrogram
         mel = acoustic_loaded.generate(units).transpose(1, 2)
+        progress(0.7, desc="Generating audio waveform")
+
         # Generate audio waveform
         target = hifigan_loaded(mel)
         progress(0.9, desc="Postprocessing audio")
@@ -86,7 +90,7 @@ def gui():
 
         with gr.Row():
             with gr.Column():
-                audio_input = gr.Audio(sources=["upload", "microphone"], type="filepath", label="Upload Audio")
+                audio_input = gr.Audio(sources=["upload", "microphone"], type="filepath", label="Upload Audio", autoplay=True, editable=False)
                 convert_button = gr.Button("Convert Speech", interactive=False)
                 info = gr.Markdown("", visible=False)